A C I L A L L E N C O N S U L T I N G

REPORT TO
RET REVIEW EXPERT PANEL

7 AUGUST 2014

RET REVIEW
MODELLING

MARKET MODELLING OF
VARIOUS RET POLICY
OPTIONS
 ACIL ALLEN CONSULTING PTY LTD
ABN 68 102 652 148

LEVEL FIFTEEN
127 CREEK STREET
BRISBANE QLD 4000
AUSTRALIA
T+61 7 3009 8700
F+61 7 3009 8799

LEVEL TWO
33 AINSLIE PLACE
CANBERRA ACT 2600
AUSTRALIA
T+61 2 6103 8200
F+61 2 6103 8233

LEVEL NINE
60 COLLINS STREET
MELBOURNE VIC 3000
AUSTRALIA
T+61 3 8650 6000
F+61 3 9654 6363

LEVEL ONE
50 PITT STREET
SYDNEY NSW 2000
AUSTRALIA
T+61 2 8272 5100
F+61 2 9247 2455

SUITE C2 CENTA BUILDING

118 RAILWAY STREET
WEST PERTH WA 6005
AUSTRALIA
T+61 8 9449 9600
F+61 8 9322 3955

ACILALLEN.COM.AU

RELIANCE AND DISCLAIMER

THE PROFESSIONAL ANALYSIS AND ADVICE IN THIS REPORT HAS BEEN PREPARED BY ACIL ALLEN CONSULTING FOR
THE EXCLUSIVE USE OF THE PARTY OR PARTIES TO WHOM IT IS ADDRESSED (THE ADDRESSEE) AND FOR THE
PURPOSES SPECIFIED IN IT. THIS REPORT IS SUPPLIED IN GOOD FAITH AND REFLECTS THE KNOWLEDGE,
EXPERTISE AND EXPERIENCE OF THE CONSULTANTS INVOLVED. THE REPORT MUST NOT BE PUBLISHED, QUOTED
OR DISSEMINATED TO ANY OTHER PARTY WITHOUT ACIL ALLEN CONSULTING’S PRIOR WRITTEN CONSENT. ACIL
ALLEN CONSULTING ACCEPTS NO RESPONSIBILITY WHATSOEVER FOR ANY LOSS OCCASIONED BY ANY PERSON
ACTING OR REFRAINING FROM ACTION AS A RESULT OF RELIANCE ON THE REPORT, OTHER THAN THE ADDRESSEE.
IN CONDUCTING THE ANALYSIS IN THIS REPORT ACIL ALLEN CONSULTING HAS ENDEAVOURED TO USE WHAT IT
CONSIDERS IS THE BEST INFORMATION AVAILABLE AT THE DATE OF PUBLICATION, INCLUDING INFORMATION
SUPPLIED BY THE ADDRESSEE. UNLESS STATED OTHERWISE, ACIL ALLEN CONSULTING DOES NOT WARRANT THE
ACCURACY OF ANY FORECAST OR PROJECTION IN THE REPORT. ALTHOUGH ACIL ALLEN CONSULTING EXERCISES
REASONABLE CARE WHEN MAKING FORECASTS OR PROJECTIONS, FACTORS IN THE PROCESS, SUCH AS FUTURE
MARKET BEHAVIOUR, ARE INHERENTLY UNCERTAIN AND CANNOT BE FORECAST OR PROJECTED RELIABLY.
ACIL ALLEN CONSULTING SHALL NOT BE LIABLE IN RESPECT OF ANY CLAIM ARISING OUT OF THE FAILURE OF A
CLIENT INVESTMENT TO PERFORM TO THE ADVANTAGE OF THE CLIENT OR TO THE ADVANTAGE OF THE CLIENT TO
THE DEGREE SUGGESTED OR ASSUMED IN ANY ADVICE OR FORECAST GIVEN BY ACIL ALLEN CONSULTING.

© ACIL ALLEN CONSULTING 2014

Executive summary
The Commonwealth Government has appointed an Expert Panel to conduct the 2014
review of the Renewable Energy Target (RET). The Expert Panel is supported by a
secretariat in the Department of the Prime Minister and Cabinet (the Secretariat).
The RET is comprised of two separate, but related schemes, namely: the Large-scale
Renewable Energy Target (LRET) and the Small-scale Renewable Energy Scheme (SRES).
ACIL Allen Consulting (ACIL Allen) has been engaged to undertake detailed electricity
market modelling of the RET impacts on Australia’s electricity markets and emissions from
electricity generation. The modelling and analysis is designed to support the Expert Panel’s
deliberations and inform the Government’s response to the Review.

Cases and sensitivities

ACIL Allen has been tasked with modelling a range of policy scenarios and sensitivities as
required by the Expert Panel. These are:
 Reference case: This case provides projections for the status quo where legislation
underpinning the LRET and SRES schemes remains unchanged and the market
develops in accordance with baseline assumptions in terms of demand and supply. All
subsequent policy scenarios are compared against this Reference case.
 Repeal case: This case assumes that the SRES and LRET schemes cease to operate
from 1 January 2015 with 2014 being the last compliance year. This scenario assumes
that any mechanism introduced to compensate investments made under the RET (if
any) does not affect wholesale or retail price outcomes.
 Closed to New Entrants: This scenario assumes that the LRET scheme continues to
operate, but is closed to new installations from 1 January 2015. The SRES, which
operates under a deeming arrangement whereby installations receive certificates
upfront, does not continue to operate and is closed from 1 January 2015. Under the
LRET, installations receive certificates annually based on generation. Closure of the
scheme to new entrants (new accredited generators) means that creation of LGCs is
limited to existing or committed generators.
 Real 20% case: The Real 20% scenario involves two significant changes to the
current policy: Reducing the LRET 2020 target to 25,500 GWh (a level which, when
evaluated using the Panel’s methodology, represents a ‘Real 20%’ of expected
demand in 2020); Closing the SRES after 2020 and reducing the period of deeming for
solar PV from 15 years down to 10 years from 1 January 2015 (deeming period is
constant at 10 years through to the end of 2020).
 Real 30% case: This scenario involves modifying the LRET target level to 30% of
anticipated demand in 2030 and extension of the scheme to 2040. Annual targets from
2015 to 2030 follow a linear trajectory, reaching 52,500 GWh and are held constant at
this level until 2040. There is no change to the current SRES, with the scheme
terminating in 2030.
 50% Growth case: This scenario was undertaken after the stakeholder workshop in
late June where preliminary modelling results were presented. It involves moving the
LRET away from fixed annual targets to floating targets with each year reset based on

ii
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
 forecast demand growth for the year ahead. The LRET target would be increased each
year based on 50% of the anticipated growth in market-facing demand; i.e. demand
growth net of that absorbed by behind the meter solar PV. The scenario assumes
SRES modifications as follows:
 A reduction in deeming for Small Generating Units (SGUs) to 10 years from 1
January 2015, with the deeming period for both SGUs and Solar Water Heaters
(SWHs) declining by one year each year and the scheme terminating at the end of
2020
 A reduction in the maximum size eligibility of small generating units for inclusion
under SRES down from the current 100 kW to 20 kW (systems above 20 kW would
be eligible for the LRET).
Figure ES 1 summarises the LRET annual targets across the policy cases. In the 50%
Growth case, the LRET annual targets are a function of demand growth and therefore vary
across the demand sensitivities examined. The Real 30% case also includes an extension
of the scheme to 2040 with targets held constant at the 2030 level until 2040 (not shown in
the figure). Table ES 1 summarises the SRES settings under each policy scenario.

Figure ES 1 LRET annual targets under the various policy scenarios

60,000

50,000

40,000
GWh

30,000

20,000

10,000

0
2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030

Reference case Real 20% case Real 30% case

50% Growth case 50% Growth case: Low Demand 50% Growth case: High demand

Note: Under all scenarios the LRET terminates in 2030 except for the Real 30% which extends out to 2040
Source: ACIL Allen based on input settings provided by the Expert Panel

iii
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
Table ES 1SRES settings under the various policy scenarios
Policy scenario Scheme end Treatment of SGU Treatment of SWH

15 years upfront, with deeming 10 years deeming upfront, with

Reference case End of calendar year 2030 period declining by 1 year each deeming period declining by 1
year from 2017 year each year from 2022

Repeal case 2014 last compliance year No further subsidies No further subsidies
Closed to new entrants case 2014 last compliance year No further subsidies No further subsidies

10 years deeming from 1

Real 20% case End of calendar year 2020 January 2015 (10 years available No change to Reference case
until scheme end)

Real 30% case End of calendar year 2030 No change to Reference case No change to Reference case

10 years from 1 January 2015, 10 years from 1 January 2015,

50% Growth case End of calendar year 2020 with the deeming period with the deeming period
declining by one year each year declining by one year each year

Source: ACIL Allen based on input settings provided by the Expert Panel

Input assumptions for the modelling have been sourced from a range of publicly available
sources including AEMO and IMO for demand and BREE for capital costs and learning
rates. These have been supplemented by ACIL Allen’s own in-house assumptions for other
key inputs. Sensitivities have also been completed to test the effects of changes for a
number of the key input assumptions where they are subject to considerable uncertainty.
These include high and low demand growth; the potential introduction of other abatement
policies modelled through a shadow carbon price from 2021; high capital costs for
renewable energy technologies; and permanent retirements for incumbent generators which
mothball capacity.

Analysis and findings

Currently renewable generation accounts for an estimated 16.1% of generation (at the end
of calendar year 2014). Under the Reference case where the RET remains unchanged,
renewable energy is projected to reach 26.3% by 2020 as shown in Figure ES 2.1
Under the Reference case assumptions, ACIL Allen’s modelling projects the renewable
energy target can be met by new renewable developments with the LRET fully subscribed
throughout the period to 2030. Much of the anticipated large-scale renewable development
occurs over the period 2016 to 2021, with around 7,650 MW of wind developed throughout
the NEM and SWIS regions and around 1,400 MW of utility-scale solar PV developed in the
regional grids of the NWIS, DKIS and Mt Isa. Owing to the subdued demand conditions in
electricity markets, the introduction of large volumes of renewable capacity results in a
mothballing of generating plant by incumbent operators, with much of this capacity returning
to service over time as demand grows.

1 This assessment has been undertaken using a formula provided by the Expert Panel and excludes the displacement from
solar water heaters (SWH). If displacement from SWH was to be added to both the renewable energy component (the
numerator), and to aggregate electricity demand (the denominator), aggregate renewables would be around one
percentage point higher at 27.3% by 2020.

iv
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
Figure ES 2 Proportion of renewables in Australia’s energy mix: Reference case

80,000 28%
26.3% 26.3% 26.3% 26.4% 26.4% 26.4% 26.4% 26.3% 26.3% 26.2% 26.1%

70,000 26%
23.6%
24%
60,000

Share of Australian energy

21.7%
22%
50,000 19.7%
GWh

20%
40,000 17.9%
17.0% 18%
16.1%
30,000
16%
20,000
14%

10,000 12%

0 10%
2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Calendar year
Existing baseline generators SRES solar PV generation
Current LRET target Proportion of renewables

Note: Proportion of estimated total Australian electricity demand

Source: ACIL Allen

Across Australia, a total of $26.8 billion (real 2014 dollars) or $15.9 billion (in present value
terms) in capital expenditure on new generating capacity is projected to occur over the
period to 2040. Wind investment is projected to account for around 62% ($16.4 billion in real
2014 dollars or $12.1 billion in present value terms) of new large-scale generation
investment in the period to 2040. Gas-fired peaking plant and utility-scale solar PV each
account for around 11% of the total ($3 billion in real 2014 dollars). In present value terms,
solar accounts for $1.8 billion compared with $0.7 billion for peaking gas as adoption times
differ. Several categories of fossil fuel generation collectively account for the remaining 16%
($4.3 billion in real 2014 dollars or $1.3 billion in present value terms).
Small-scale systems (solar PV and solar water heaters) under the SRES are projected to
see strong growth with solar PV capacity rising from 4,133 MW at the end of 2014 to just
under 13,000 MW by 2030. Cumulative SWH installations are projected to increase from an
estimated 915,000 at the end of calendar year 2014 to over 1.5 million systems by 2030. A
total of $30.4 billion (real 2014 dollars) or $18 billion (present value terms) of new
investment is projected to occur in relation to solar PV and SWHs over the period to 2030.
The majority of this is solar PV ($20.6 billion in real 2014 dollars or $12.6 billion in present
value terms).
However, the subsidies paid to the renewable energy industry through the RET to bring
about this investment are high. Over the period to 2030, the projected total direct RET cost
(projected number of certificates multiplied by price) is $37.8 billion in real 2014 dollars
($22.4 billion in present value terms), of which, over 80% is associated with the LRET.
The modelling also shows that much of this additional capacity developed under the LRET is
surplus to market needs. Under the Repeal scenario, the modelling projects a net reduction
in the development of generating capacity of around 8,500 MW. Given the current levels of
oversupply in most electricity grids and muted demand growth, the existing generation fleet
is almost sufficient to meet expected demand for the foreseeable future.

v
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
 Generation sector resource costs
Figure ES 3 below presents a summary of the present value of aggregate generation sector
resource costs over the period 2015-40 across each of the scenarios and sensitivities
modelled. This measure can be interpreted as the cost to society of generating electricity for
consumption by consumers and provides an indication of the sector’s labour and capital
productivity under each scenario and sensitivity when viewed on a per MWh basis.
For the Reference case (under core assumptions) costs total $121.9 billion in present value
terms over the period to 2040 using a discount rate of 7% pre-tax real. Under the core
assumptions, all of the policy variants examined resulted in a reduction in sector resource
costs, indicating capital and/or labour productivity gains for the economy. The Real 30%
scenario has almost the same aggregate cost as the Reference case because the deferral
of wind development early is offset by an overall larger amount of renewable development in
the longer-term.
The Repeal case has the lowest projected resource costs, as expected, as there are no
RET subsidies distorting supply costs and competitive wholesale electricity markets are left
to determine the most efficient, least cost plant mix to meet demand. This was one of the
fundamental intentions in the establishment of the NEM, with its rules and principles being
deliberately technology agnostic. Another reason for the development of the NEM was to
impose competitive disciplines on participants in order to avoid the large oversupply in
generation that had occurred through state governments using electricity supply to support
other industries and policies. In a market with little or no demand growth, the RET is creating
the same oversupply in generation that the NEM was designed to correct. In the absence of
the RET policy, the market determines the optimal level of generation investment, rather
than having arbitrary targets imposed upon it. In a market environment where capacity is
already oversupplied and demand may continue to decline, it is desirable (and efficient) for
no new investment in capacity to occur.

Figure ES 3 Aggregate generation sector resource costs (NPV 2015-2040): All scenarios/sensitivities
132.9

138.1
126.7
123.8

123.2
122.8
121.9
121.8

160
112.7

112.5

111.8
111.4

108.9
108.6
108.2

108.0
107.9

106.4
106.0

140
96.0
90.8
90.7
Real 2014 $ billion

120
100
80
60
40
20
0
Real 20%

Real 30%

Real 20%

Real 30%
Repeal

Repeal

Repeal

Repeal

Repeal
Closed to New Entrants

Closed to New Entrants

50% Growth

50% Growth

50% Growth

50% Growth

50% Growth
Reference

Reference

Reference

Reference

Reference

Reference

Core Carbon High capital costs High demand Low demand Permanent
retirements

Note: Measure includes capital expenditure (on both generating capacity and any interconnector expansions/augmentations); refurbish ment of
existing and new generators for life extension beyond initial economic life; fixed operating costs (fixed costs associated with normal operation
and stay in business capital expenditure associated with existing and new generating capacity); variable operating costs (fuel costs and variable
O&M costs for existing and new generation) and unserved energy. NPV calculated using a 7% real discount rate.
Source: ACIL Allen

vi
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
 Emissions and cost of abatement
The RET policy delivers emissions abatement through displacing fossil fuel based
generation with renewable generation. The level of abatement achieved is projected to be
higher under the current market conditions relative to previous assessments because of the
reduced role of gas-fired plant (increasing gas prices) and the repeal of the carbon price,
both of which increase the competitiveness of coal fired plant within the generation mix.
Figure ES 4 below shows that the policy scenarios which include the RET or an expansion
to the RET (the Real 30% case) consistently result in the lowest emissions outcomes across
assumption sets. Conversely the Repeal of the RET is projected to lead to higher emissions;
between 8% and 14% relative to the Reference case over the period to 2040.

Figure ES 4 Aggregate emissions from electricity generation: 2015-2040: All scenarios/sensitivities

5,784
5,481
5,345
7000

5,117
5,114

5,114

5,072

5,072
4,917

4,917

4,909

4,859
4,594

4,593
4,557
4,525

4,494
6000

3,973
3,952
3,781
3,463
3,393
5000
4000
Mt CO2-e

3000
2000
1000
0
Real 30%

Real 20%

Real 30%

Real 20%
Repeal

Repeal

Repeal

Repeal

Repeal
Closed to New Entrants

Closed to New Entrants

50% Growth

50% Growth

50% Growth

50% Growth

50% Growth
Reference

Reference

Reference

Reference

Reference

Reference
Core Carbon High capital costs High demand Low demand Permanent
retirements

Note: Excludes non-scheduled generation in NEM regions, own-generation in the SWIS and off-grid generation ‘Other’ category includes
cogeneration, liquid fuels, CCS-equipped technologies, biomass and geothermal.
Source: ACIL Allen

As with any Government expenditure or program, an important consideration is whether the

policy offers value for money relative to alternatives. Two methods for calculating the cost of
abatement from the policy have been used. Method 1 calculates abatement costs as the
present value of the change in resource costs divided by the discounted change in
abatement. Method 2 is the same, except the emissions in the denominator are not
discounted.2 Using method 1, estimated abatement costs for the RET range from $59/tonne
under the 50% growth case (core assumptions) to $77/tonne (Reference case High capital
costs).3 Under method 2, these costs are lower, ranging from $30/tonne to $40/tonne under
the same scenarios. Whilst the policy is somewhat effective in the abatement of emissions,
it is at high cost compared to current global pricing and is therefore not the most efficient
means of emissions abatement. There is also a large difference between calculated
abatement costs for the LRET and SRES components, with the abatement costs for the
SRES being at least 2 to 3 times higher than the LRET. Therefore policy scenarios which
tend to reduce subsidies provided for solar PV will tend to lower the overall RET abatement
cost.

2 ACIL Allen has used the second method to calculate abatement costs at the request of the Expert Panel. However, ACIL
Allen considers that the second method does not appropriately reflect the costs of emissions abatement on an inter-
temporal basis.
3 This range excludes the 50% Growth low demand case which is an outlier with a much higher cost of $164/tonne.

vii
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
Impacts on retail prices
The public analysis of the costs and benefits of the RET scheme has been dominated by
views on the net benefits that the RET scheme provides to electricity consumers. These net
electricity consumer benefits are generally calculated by assessing wholesale electricity
price (pool) and RET certificate price changes for the market with and without the RET
scheme (i.e. the modelled impact of the subsidised renewable generation on wholesale
electricity market prices). Assessing the net consumer benefits limited to a specific
economic sector cannot be considered to be either a social cost benefit analysis or an
economy wide assessment of the RET scheme. Considering only the benefits that flow to
consumers ignores the opportunity costs of the capital and labour involved and the other
welfare effects of the policy.
Figure ES 5 below shows the projected aggregate cost for an average Australian household
on electricity over the period 2015-40 in NPV terms. In most cases, moving from the
Reference case to the Repeal case (the most extreme policy variant) results in projected
household electricity costs rising in net terms (the reduction in direct compliance costs is
outweighed by increases in wholesale electricity prices). This indicates that wealth transfers
are occurring from existing generators to both new renewable energy projects and
consumers.
Interestingly, this pattern of price changes does not hold under low demand conditions. This
is because new renewable generation is incapable of supressing wholesale prices below
levels which are sustainable for incumbent generators to keep operating. Under these
conditions, removal of the direct compliance costs is not offset by any increases in
wholesale prices and consumers are better off under a Repeal scenario.
The impact on retail electricity prices is subject to uncertainty in the modelled components.
Pool prices are inherently uncertain. This is because many of the drivers of pool prices are
uncertain, such as:
 Weather driving demand is unpredictable and highly variable
 Plant performance (outages) is also stochastic (random)
 Fuel prices may vary over time although most fossil fuel fired plant tend to contract over
several years and so these prices tend to be reasonably certain on an annual basis
 Participant behaviour (mothballing, plant retirement, strategic bidding, etc.) may swamp
other effects over time.
A key factor in the uncertainty around future electricity prices is participant behaviour. As
electricity demand has fallen in recent years, an increasing willingness of participants to
mothball or close generation plant has been observed. Closing or mothballing plant can
cause a significant rebound in pool prices and may fully offset any downward pressure from
renewable plant. While we have incorporated some mothballing of plant in the analysis,
participants may have different objectives and take quite different views to mothballing and
plant closure than we have taken. This could substantially change the net benefit to
electricity consumers through net changes in retail prices. The Permanent retirement
sensitivity, which includes a larger amount of incumbent capacity withdrawal, demonstrates
that the impact on retail electricity prices can easily be reversed, with consumers benefiting
from moving from the Reference case policy to a 50% Growth scenario. Directionally, the
same outcome would also be seen if the scheme was closed to new entrants or fully
repealed.

viii
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
Figure ES 5 NPV of average household total expenditure on electricity (2015-2040): All
scenarios/sensitivities

20,260

20,451
20,074

20,037
21,000

19,887

19,771
19,754

19,634

19,464
19,453
20,500

19,358
19,349
19,305

19,213
19,193
19,182
19,181
19,119

19,119
19,092
19,003
20,000
NPV of retail bills ($)

18,706
19,500
19,000
18,500
18,000
17,500
Real 30%

Real 20%

Real 30%

Real 20%
Repeal

Closed to New Entrants

Repeal

Closed to New Entrants

Repeal

Repeal

Repeal
Reference

50% Growth

Reference

Reference

50% Growth

Reference

50% Growth

50% Growth

Reference

50% Growth

Reference
Core Carbon High capital costs High demand Low demand Permanent
retirements

Note: NPV of annual residential bills for average household over the period 2015-40. Uses a 7% real discount rate
Source: ACIL Allen

Regardless of direction, the impact on retail electricity prices is small, even when considered
over the period to 2040. Under the core assumptions, moving from the Reference case to a
complete repeal of the scheme is projected to increase a typical household’s expenditure on
electricity over the period to 2040 by 0.6% in present value terms. By comparison, moving
from the Reference case to a Repeal case under low demand conditions is projected to
reduce a typical household’s expenditure on electricity (over the same period) by 2.1% in
present value terms. In all cases examined, the benefits or costs are a very small
percentage of the total electricity bill and could easily be swamped by the range of
uncertainties in pool prices, especially the changes in the behaviour of generation
participants.
Assessing the RET’s impacts on retail electricity prices in isolation does not provide a solid
basis for economically evaluating the RET policy. That the RET may lower electricity prices
for consumers does not mean that its benefits outweigh its costs when considered in society
wide terms. The diversion of capital and labour from other productive activities to the
electricity sector imposes real costs on other sectors of the economy. Other policies such as
subsidising fossil fuels or fossil fuel generators would also likely have the effect of lowering
costs to electricity consumers and probably at significantly lower resource costs, yet few
would advocate these as being good policy positions. An economic evaluation of the policy
would not normally include wealth transfers where either producers or consumers benefit at
the expense of each other. This makes projected changes to retail electricity prices mostly
irrelevant in any economic assessment of the policy.
In ACIL Allen’s view, the main focus of any evaluation should be on the cost of abatement
achieved through the policy and whether this represents an efficient means of achieving
abatement objectives.

ix
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS